CN107061996A - A kind of water supply line leakage detecting and locating method - Google Patents
A kind of water supply line leakage detecting and locating method Download PDFInfo
- Publication number
- CN107061996A CN107061996A CN201710139772.7A CN201710139772A CN107061996A CN 107061996 A CN107061996 A CN 107061996A CN 201710139772 A CN201710139772 A CN 201710139772A CN 107061996 A CN107061996 A CN 107061996A
- Authority
- CN
- China
- Prior art keywords
- signal
- leakage
- noise
- spectrum
- leak
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L21/0232—Processing in the frequency domain
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
Abstract
The present invention provides a kind of water supply line leakage detecting and locating method, the signal that methods described is gathered to sensor carries out spectrum and subtracts enhancing, calculate the Spectral variance of signal after enhancing, judged using double threshold method, if Spectral variance is in threshold range there is leakage in explanation, leak point location is carried out again, wave filter is constituted using BP neural network, water leakage is separated from noise and generalized correlation is carried out, the good weight function of performance is selected to carry out time delay estimation according to signal to noise ratio, obtain the time delay of three sensors, calculated using leak location model, obtain leak dot position information.The present invention, to signal enhancing, then carries out leakage judgement using spectrum-subtraction using double threshold method, and explanation has leakage if Spectral variance is in threshold range, and accuracy is higher;Estimate that lifting of the noise spectrum to signal to noise ratio is more obvious using " the No leakage estimation technique ", be filtered using BP neural network, the Time delay Estimation Accuracy of generalized correlation improved, so as to effectively improve leak spot placement accuracy.
Description
Technical field
The invention belongs to leak localization technical field, and in particular to a kind of water supply line leakage detecting and locating method.
Background technology
Water supply pipe leakage detection and localization is the important prerequisite that pipe-line maintenance is repaired.According to official statistics, the country more than 600
The average leak rate of individual public supply mains is more than 20%, and up to more than 60%.China's most area is only in running water
Large area is leaked, or even leakage can be just judged in the case of pipeline burst.Therefore, detect in time and orient leakage hair
Raw point is significant for the saving of water resource.
Pipe leakage signal in practice is mixed by leakage signal and various ambient noises.Leakage signal can recognize
To be stationary signal, the ground that ambient noise includes the caused pipe vibration noise of running water flowing, motor-driven vehicle going is caused shakes
Noise etc. caused by dynamic, construction, because external interference noise not necessarily meets smooth conditions, can bring greatly tired to leak detection
It is difficult.
Therefore, first of all for leak detection effect when being lifted at low signal-to-noise ratio, the present invention is with leakage signal and noise
The difference of spectrum signature is starting point, it is proposed that the leak detection algorithm based on spectrum-subtraction and Spectral variance.Spectrum-subtraction is a kind of
Effective voice enhancement algorithm, its algorithm complex is low, real-time, and the present invention is applied in the detection to leakage signal
On.First, the frequency spectrum of noise is estimated using the algorithm, then is strengthened water leakage by " spectrum subtraction ".Frequency spectrum
Variance is differed greatly using the spectral characteristic of leakage signal, and the spectral characteristic difference of ambient noise is smaller, so as to identify
Leakage signal.Next, the positioning precision in order to reduce time delay evaluated error and then raising algorithm, it is proposed that one kind is based on BP god
Leak positioning mode through network, using error back propagation mechanism, according to output valve and desired value calculation error, then by error Lai
Successively change weights so that error is minimum, so as to complete network training, constructs the filtering system based on BP neural network, enters
And obtain accurate time delay using generalized correlation and estimate, realize being accurately positioned for leakage point.
On the basis of parser principle, emulation experiment is carried out under the conditions of different signal to noise ratio.Test result indicates that, should
Method can also obtain preferably detection locating effect under low signal-to-noise ratio and environmental condition complicated and changeable.
The content of the invention
In order to solve the above problems, the present invention provides a kind of water supply line leakage detecting and locating method, and methods described is to passing
The signal of sensor collection is strengthened, and is then calculated the Spectral variance of signal after enhancing, is judged using double threshold method, if frequency
Spectrum variance is in threshold range then explanation and there is leakage, then carries out leak point location, and wave filter is constituted using BP neural network,
Water leakage is separated from noise, generalized correlation is carried out to water leakage, is entered according to the weight function that signal to noise ratio selects performance good
Row time delay estimates, obtain three sensors when delay, calculated using leak location model, obtain leakage point position and believe
Breath;
Further, methods described includes:
S1:Water leakage progress spectrum is subtracted into enhancing;
S2:Calculate the Spectral variance of signal after strengthening in S1;
S3:Leakage signal is recognized using one-parameter double-threshold comparison, and determines threshold value;
S4:Leak detection is carried out, and exports testing result;
S5:To the testing result exported in S4, out-of-band noise is removed by bandpass filtering;
S6:NN filtering is carried out, in-band noise is removed;
S7:Carry out generalized correlation;
S8:Result to generalized correlation in S7 carries out time delay estimation, and is calculated by leak location model, is leaked
Water spot positional information, completes leak positioning;
Further, the S7 is specially:During selection of weighting function, the good power letter of performance is selected according to signal to noise ratio
Number carries out time delay estimations, obtain three sensors when delay, calculated using the leak location model, obtain leakage point
Positional information;
Further, the leak location model is three sensor location models, and the model is by equidistant point of sensor
Cloth is around water supply network, wherein the three sensor nodes numbering nearest apart from leak source is respectively 1,2,3, between them away from
From for L, sensor 1,2 is set to d1 and d2 apart from the position of leak source, is measured by correlation method between sensor 1 and 2,2 and 3
Time delay be respectively D12And D23, the spread speed ν where water leakage on pipeline:ν=L/D23, leak source is away from sensor 1 and 2
Distance and be respectivelyWith
Further, the calculating of spectrum-subtraction is specific as follows in the S1:
S11:Input water leakage s (n) is pre-processed, the pretreatment includes preemphasis and adding window framing;
S12:The leakage signal made an uproar to band carries out Short Time Fourier Analysis, calculates the short-time energy spectrum for obtaining each frame signal
|Ym(ω)|2;
S13:Before strengthening using spectrum-subtraction water leakage, noise spectrum estimation is carried out, noise spectrum estimation is obtained
ValueWith | Ym(ω)|2SubtractSo as to obtain removing the leakage signal power spectrum after Noise enhancement
S14:Spectrum amplitude value to obtaining leakage signal after result sqrt in S13With reference to former water leakage
Phase information obtain the spectrum estimation of each frame leakage signalInverse Fourier transform is carried out again, leakage signal is carried out extensive
Multiple and reconstruct, obtains spectrum and subtracts enhanced leakage signal;
Further, the calculating of signal spectrum variance is specific as follows in the S2:
S21:Assuming that input signal is s (n), the length per frame is N, and signal is transformed from the time domain into frequency domain meter by DFT
Calculate spectrum value:
Record each frequency component with a matrix | S (ω) | value;
S22:Calculate the average of each component:
S23:Calculate the Spectral variance value of enhanced leakage signal in previous step:And
Obtain the average value of noise model Spectral variance
Further, the S3 is specially:
S31:Two threshold values T1 and T2 are set, the average value of noise model Spectral variance is obtained
S32:Threshold value T1 is set toThreshold value T2 is set to
Further, the S4 is specially:
S41:It is leakage signal when parameter D is higher than threshold value T2;
S42:T1 positions are higher or lower than by D, to judge the terminal of leakage signal;
S43:Statistics, if frame number is more than a quarter of input signal totalframes, is sentenced higher than the signal frame number of threshold value
It is set to leakage, and output signal, is otherwise No leakage;
Further, during carrying out time delay estimation using neutral net in the S6, its process is divided into two stages:
First stage is learning process, and second stage is the course of work;
Further, the learning process is:
1) sample data under selection leakage environment is used as training sample;
2) sample data is pre-processed, before neural network prediction is carried out, initial data is normalized place
Reason, makes its data standard between [- 1,1];
3) training sample is constructed, the signal sample sequence after noise suppressed is as object vector under experimental conditions,
The water leakage containing different noises is measured under actual conditions, obtained sample sequence is as input signal;
Further, the course of work is:
A) BP neural network for choosing three-decker sets up forecast model;The corresponding node of input layer, hidden layer, output layer
Number is respectively 1,40 and 1, and hidden layer output layer functions are respectively tansig and purelin functions;
B) training network;Before training network, in addition it is also necessary to set training parameter, wherein maximum training pace
Net.trainParam.epochs is 10000, and least mean-square error net.trainParam.goal is 0.05, and study speed
Rate is set to 0.01 etc.;
C) after neural network completion, the water leakage collected is carried out in band, out-of-band noise suppresses;
Beneficial effects of the present invention are as follows:
1) judged using double threshold method, explanation has leakage, accuracy if Spectral variance is in threshold range
It is higher;
2) estimate that lifting of the noise spectrum to signal to noise ratio is more obvious using " the No leakage estimation technique ";
3) leakage signal made an uproar using spectrum-subtraction to band carries out noise reduction process, effectively improves the signal to noise ratio meeting of water leakage
Contribute to the lifting of leak detection validity;
4) using noise and the difference of leakage signal spectral characteristic, the frequency spectrum of noise is evenly distributed in each frequency point
Section, and each frequency component is smaller, the value for calculating variance to its frequency spectrum is also smaller, and the spectral fluctuations of leakage signal are larger, to it
The value for calculating Spectral variance is also big.Therefore, the accuracy rate of leak detection can be effectively improved using threshold value diagnostic method;
5) in groundwater supply environment complicated and changeable, high-precision time delay estimation is difficult to realize.Propose a kind of based on BP
The leak localization method of neutral net.It is located in advance by the study of neutral net to the water leakage under varying environment
Reason, then carries out accurate time delay estimation using generalized correlation method, improves the positioning precision of leakage point.
Brief description of the drawings
Fig. 1 is leak location model of the present invention;
Fig. 2 is the algorithm flow chart of leak detection of the present invention;
Fig. 3 is based on neutral net positioning mode illustraton of model to be of the present invention;
Fig. 4 is the overall framework figure of the method for the invention;
Fig. 5 is the original and plus water leakage oscillogram of making an uproar when the method for the invention is verified;
Fig. 6 be the method for the invention verify when under different signal to noise ratio short-term spectrum variogram;
Fig. 7 is the original and filtered waveform and spectrogram that the method for the invention obtains rush-hour;
Fig. 8 is neural metwork training error curve of the present invention;
Fig. 9 is the network training under the conditions of white noise of the present invention and time delay estimation condition;
Figure 10 is the network training under the conditions of coloured noise of the present invention and time delay estimation condition;
Figure 11 is neural metwork training precision under different signal to noise ratio of the present invention;
Figure 12 is each generalized related function image when SNR of the present invention is 5dB;
Figure 13 is each generalized related function image when SNR of the present invention is -5dB.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is explained in further detail.It should be appreciated that specific embodiment described herein is used only for explaining the present invention, and
It is not used in the restriction present invention.On the contrary, the present invention cover it is any be defined by the claims the present invention spirit and scope on do
Replacement, modification, equivalent method and scheme.Further, in order that the public has a better understanding to the present invention, below to this
It is detailed to describe some specific detail sections in the detailed description of invention.It is thin without these for a person skilled in the art
The description of section part can also understand the present invention completely.
The invention will be further described with specific embodiment below in conjunction with the accompanying drawings, but not as a limitation of the invention.
Below most preferred embodiment is enumerated for the present invention:
As shown in Fig. 1-Figure 13, the present invention provides a kind of water supply line leakage detecting and locating method, and methods described is to sensing
The signal of device collection is strengthened, and is then calculated the Spectral variance of signal after enhancing, is judged using double threshold method, if frequency spectrum
Variance is in threshold range then explanation and there is leakage, then carries out leak point location, and wave filter is constituted using BP neural network, will
Water leakage is separated from noise, and generalized correlation is carried out to water leakage, selects the good weight function of performance to carry out according to signal to noise ratio
Time delay estimates, obtain three sensors when delay, calculated using leak location model, obtain leak dot position information,
Methods described includes:
S1:Water leakage progress spectrum is subtracted into enhancing;
S2:Calculate the Spectral variance of signal after strengthening in S1;
S3:Leakage signal is recognized using one-parameter double-threshold comparison, and determines threshold value;
S4:Leak detection is carried out, and exports testing result;
S5:To the testing result exported in S4, out-of-band noise is removed by bandpass filtering;
S6:NN filtering is carried out, in-band noise is removed;
S7:Carry out generalized correlation for time delay estimation;
S8:Using the time delay estimated result in S7, by leak location model, leak dot position information is obtained, leakage is completed
Water is positioned.
Described 8 are specially:During selection of weighting function, the good weight function of performance is selected to carry out time delay according to signal to noise ratio
Estimation, obtain three sensors when delay, calculated using the leak location model, obtain leak dot position information.
The leak location model is three sensor location models, and the model equidistantly distributes sensor in feed pipe
Net week is enclosed, wherein the three sensor nodes numbering nearest apart from leak source is respectively 1,2,3, the distance between they are L, sensing
Device 1,2 is set to d1 and d2 apart from the position of leak source, and the time delay measured by correlation method between sensor 1 and 2,2 and 3 is distinguished
For D12And D23, the spread speed ν where water leakage on pipeline:ν=L/D23, distance and difference of the leak source away from sensor 1 and 2
For
The calculating of spectrum-subtraction is specially in the S1:
S11:Input water leakage s (n) is pre-processed, the pretreatment includes preemphasis and adding window framing;
S12:The leakage signal made an uproar to band carries out Short Time Fourier Analysis, calculates the short-time energy spectrum for obtaining each frame signal
|Ym(ω)|2;
S13:Before strengthening using spectrum-subtraction water leakage, noise spectrum estimation is carried out, noise spectrum estimation is obtained
ValueWith | Ym(ω)|2SubtractSo as to obtain removing the leakage signal power spectrum after Noise enhancement
S14:Spectrum amplitude value to obtaining leakage signal after result sqrt in S13With reference to former water leakage
Phase information obtain the spectrum estimation of each frame leakage signalInverse Fourier transform is carried out again, leakage signal is carried out extensive
Multiple and reconstruct, obtains spectrum and subtracts enhanced leakage signal;
The calculating of signal spectrum variance is specific as follows in the S2:
S21:Assuming that input signal is s (n), the length per frame is N, and signal is transformed from the time domain into frequency domain meter by DFT
Calculate spectrum value:
Record each frequency component with a matrix | S (ω) | value;
S22:Calculate the average of each component:
S23:Calculate the Spectral variance value of enhanced leakage signal in previous step:And
Obtain the average value of noise model Spectral variance
The S3 is specially:
S31:Two threshold values T1 and T2 are set, the average value of noise model Spectral variance is obtained
S32:Threshold value T1 is set toThreshold value T2 is set to
The S4 is specially:
S41:It is leakage signal when parameter D is higher than threshold value T2;
S42:T1 positions are higher or lower than by D, to judge the terminal of leakage signal;
S43:Statistics, if frame number is more than a quarter of input signal totalframes, is sentenced higher than the signal frame number of threshold value
It is set to leakage, and output signal, is otherwise No leakage.
During being filtered in the S6 using neutral net, its process is divided into two ranks
Section:First stage is learning process, and second stage is the course of work.
The learning process is:
1) sample data under selection leakage environment is used as training sample;
2) sample data is pre-processed, before neural network prediction is carried out, initial data is normalized place
Reason, makes its data standard between [- 1,1];
3) training sample is constructed, the signal sample sequence after noise suppressed is as object vector under experimental conditions,
The water leakage containing different noises is measured under actual conditions, obtained sample sequence is as input signal.
The course of work is:
A) BP neural network for choosing three-decker sets up forecast model;The corresponding node of input layer, hidden layer, output layer
Number is respectively 1,40 and 1, and hidden layer, output layer functions are respectively tansig and purelin functions;
B) training network;Before training network, in addition it is also necessary to set training parameter, wherein maximum training pace
Net.trainParam.epochs is 10000, and least mean-square error net.trainParam.goal is 0.05, and study speed
Rate is set to 0.01 etc.;
C) after neural network completion, the water leakage collected is carried out in band, out-of-band noise suppresses.
The present invention provides a kind of water supply line leakage detecting and locating method positioning, and methods described is based on system model, and this is
System model is as shown in figure 1, sensor is equidistantly distributed around water supply network, wherein three sensors nearest apart from leak source
Node serial number is respectively 1,2,3, and the distance between they are L, and sensor 1,2 is set to d1 and d2 apart from the position of leak source,
It is respectively D that can measure the time delay between sensor 1 and 2,2 and 3 by correlation method12And D23, it is possible thereby to first try to achieve leak letter
Spread speed ν where number on pipeline is:
ν=L/D23 (1)
Then following relational expression can be obtained according to position relationship:
Formula (1) substitution formula (2) can be obtained into distance of the leak source away from sensor 1 and 2 is respectively:
Therefore, by three sensor location models, the spread speed of pipe leakage signal can be obtained, leak source can be obtained again
The position of the end sensor of distance two, so as to realize that leak source is positioned.
Leakage signal detection method principle based on spectrum-subtraction and Spectral variance is as shown in Fig. 2 spectrum-subtraction is a kind of to letter
Number enhancing denoising efficient algorithm.Spectrum-subtraction is assuming that ambient noise signal is the additive noise of short-term stationarity, primary signal
Proposed with noise in the case of incoherent.Spectrum-subtraction has that amount of calculation is small, algorithm is simple, it is easy to accomplish and denoising effect is good
The characteristics of, using quite varied.Its effect has carried out certain filtering process equivalent in transform domain to signals with noise, so as to obtain
More pure signal spectrum.
Signal y with noisem(n) it can be expressed as:
ym(n)=sm(n)+dm(n), m=1,2 ...;N=0,1 ..., N-1 (4)
Wherein dm(n) it is noise signal, sm(n) it is pure leakage signal.M represents signal analysis frame number, and N represents letter
Number analysis frame length.Fourier transformation is done to above formula two ends simultaneously to obtain:
Ym(ω)=Sm(ω)+Dm(ω) (5)
Wherein, Ym(ω)、Sm(ω) and Dm(ω) corresponds to y respectivelym(n),smAnd d (n)m(n) carry out after Fourier transformation
Spectral density.It is squared to its, Y can be obtainedmThe short-time energy spectrum of (ω):
|Ym(ω)|2=| Sm(ω)|2+|Dm(ω)|2+Sm(ω)·Dm(ω)*+Sm(ω)*·Dm(ω) (6)
Obtained by above formula:
As it is assumed that smAnd d (n)m(n) it is independent, so Sm(ω) and Dm(ω) is also independent, and assumes Dm(ω) is zero equal
The Gaussian Profile of value, so:
It can obtain:
|Ym(ω)|2=| Sm(ω)|2+|Dm(ω)|2 (9)
Due to stationary noise frequency spectrum before and after leakage it is considered that do not change substantially, so can be by leakage before
So-called " quiet section " estimates the energy spectrum of noise | Dm(ω)|2.The purpose of Signal Enhanced Technology based on short-time spectrum amplitude Estimation
Exactly try to obtain | Sm(ω) | estimationAnd thus obtain sm(n) estimationI.e. enhanced leakage signal.
It can obtain:
It so can be obtained by enhanced leakage signal:
The gain function for defining m-th of frequency component is:
It can obtain:
Sm(ω)=Gm(ω)·Ym(ω) (13)
That is, being multiplied by a coefficient G to each spectrum component of signals with noisem(w) leakage signal after denoising, is obtained
Frequency spectrum.Using original tape make an uproar leakage signal phase spectrum come instead of estimation after signal phase spectrum.It is carried out in Fu again
Leaf inverse transformation can be obtained by the leakage signal of denoising.
Actual signal differs greatly with the spectral characteristic for being mingled in noise signal therein.The live part collection of leakage signal
In in this frequency range of 500~3000Hz, and the range value fluctuation of each frequency range is violent, and signal is steady in a short time
, it is exactly to concentrate on extremely narrow frequency range to be reflected in frequency domain, and is approximately zero in the spectrum of other frequency ranges.Therefore to leakage signal meter
Calculate Spectral variance, it will obtain a larger value.White noise is the noise being most widely present in actual applications, its power
Then relatively flat, the bandwidth of distribution is composed, and spectrum is smaller, and variance is calculated to its frequency spectrum, obtained value is smaller, and substantially small
In the Spectral variance value of leakage signal.It therefore, it can distinguish leakage signal and noise signal according to Spectral variance.
The calculating detailed process of signal spectrum variance is as follows:
(1) it is s (n) to assume input signal, and the length per frame is N.Signal is transformed from the time domain to by frequency-domain calculations by DFT
Spectrum value:
Record each frequency component with a matrix | S (ω) | value.
(2) average of each component is calculated:
(3) Spectral variance value is calculated:
Spectral variance reflection is can be seen that from formula (16) is the fluctuating quantity of each component of signal frequency domain, and frequency spectrum is with frequency
The change of rate is more violent, then D value is bigger.And for white noise signal, its spectral ripple it is shallower and distribution frequency range it is wide.Cause
This Spectral variance is smaller.Thus, it is possible to which echo signal is detected according to Spectral variance.
BP neural network is one kind with the forward direction based on error back propagation (Error Back Propagation, BP)
Network, Fig. 3 be its topological structure, it is made up of input layer, hidden layer, the part of output layer three, wherein every layer can include it is many
Individual neuron node realizes that the neuron between full connection, same layer is connectionless between layers as single input.
Neural network structure figure as shown in Figure 3, then can calculate the reality output for drawing network and desired output difference
For:
D (n)=[d1,d2,…,dJ] (18)
Then the error of nth iteration is:
ej(n)=dj(n)-Yj(n) (19)
It is defined as so obtaining error energy:
During error back propagation, according to steepest descent method, gradient of the error to weights is calculated, further along gradient
Opposite direction carry out weighed value adjusting,
According to the chain type of differential rule, the computational methods of gradient are obtained:
Introduce the concept of partial gradient:
For output layer, transmission function is generally linear function, therefore its derivative is designated as constant k, and learning rate is η,
Substitute into the correction that above formula draws neuron weights:
The key technology of leak positioning is exactly the time delay estimation of reception signal between sensor, generally, two paths of signals time
Delay can carry out peakvalue's checking by the cross-correlation function to two paths of signals and estimate.As in Fig. 1, it is assumed that what leakage point was sent
Signal is S (n), then
Wherein, the signal received by sensor 1 and 2 is respectively S1And S (n)2(n), n1And n (n)2(n) believe for noise
Number, α is decay factor, TSFor the sampling period.S1And S (n)2(n) cross-correlation function is defined as:
According to the property of stationary random process auto-correlation function, have
Then obtain, i.e., the time delay between sensor 1 and 2 is:
The present invention relates to the leakage signal detection algorithm based on spectrum-subtraction and Spectral variance, algorithm principle is as shown in Figure 2.
Spectrum is subtracted enhancing algorithm with Spectral variance to combine.Noise reduction process is carried out first by the leakage signal that spectrum-subtraction is made an uproar to band,
The signal to noise ratio for effectively improving water leakage may consequently contribute to the lifting of leak detection validity.Noise and leakage signal are utilized afterwards
The difference of spectral characteristic, the frequency spectrum of noise is evenly distributed in each frequency segmentation, and each frequency component is smaller, to its spectrometer
The value for calculating variance is also smaller.The Spectral variance of noise signal has notable difference relative to the Spectral variance value of leakage signal.Finally
By setting suitable threshold value to identify leakage signal exactly, leakage is judged.
The step of realizing of specific algorithm is:
1) input water leakage s (n) is pre-processed, including preemphasis, adding window framing.
2) leakage signal made an uproar to band carries out Short Time Fourier Analysis.Calculate the short-time energy spectrum for obtaining each frame signal |
Ym(ω)|2。
3) before strengthening using spectrum-subtraction water leakage, noise spectrum estimation is carried out first, noise spectrum is obtained and estimates
EvaluationWith | Ym(ω)|2SubtractSo as to obtain removing the leakage signal power spectrum after Noise enhancement
4) the spectrum amplitude value of leakage signal is obtained after sqrtObtained with reference to the phase information of former water leakage
To the spectrum estimation of each frame leakage signalInverse Fourier transform is carried out again, leakage signal is recovered and reconstructed, and is obtained
Subtract enhanced leakage signal to spectrum.
5) leak detection is carried out using Spectral variance, calculates enhanced each frame signal in previous stepWith noise
The Spectral variance value D of model.And obtain the average value of noise model Spectral variance
6) present invention uses one-parameter double-threshold comparison method, the only one of which parameter " frequency during detection leakage signal
Variance D " is composed, then leakage signal is recognized with double threshold.Two threshold values T1 and T2 are set, when parameter D is higher than threshold value T2
When be judged as leakage signal, then when judging the terminal of leakage signal higher or lower than T1 from D.Wherein, threshold value
T1 is set toThreshold value T2 is set to
7) in leak detection algorithm running, statistics is believed higher than the signal frame number of threshold value if frame number is more than to input
The a quarter of number totalframes, then be determined as leakage, and output signal.Otherwise it is No leakage.
Leak location algorithm principle as shown in Figure 3.The general thought of methods described is:Generalized correlation method carries out time delay estimation
As a result it is more accurate, but need a large amount of prioris and statistical property to determine the use of weight function, this is for underground leak
It is difficult to realize for detecting system, therefore neutral net positioning mode is constituted with reference to generalized correlation method based on BP neural network.
And common generalized correlation function includes:Roth processors, SCOT (smooth coherence transfer), PHAT (phse conversion),
This 5 kinds of Eckart, ML (maximum likelihood), optimizes along with traditional correlation method can obtain 6 kinds of prefilter models altogether, is returning
Receive out after 6 kinds of conventional weight functions, gather water supply line leakage when noiseless or high s/n ratio in laboratory conditions first
Water signal, then gathers the water leakage under different noise situations.Discrete sampling, structure are carried out to the water leakage under different situations
Into one-dimensional row vector matrix so that neutral net is used.In order to overcome generalized correlation method to need a large amount of this drawback of priori conditions,
Need the learning functionality and adaptivity using neutral net.After many experiments are carried out, determine to build the BP god of three-decker
Through network.
Wherein the nodes of input layer are 1, the one-dimensional row vector matrix that input content is made up of water leakage sampled point.
Hidden layer is from tansig functions as excitation function, and the nodes of hidden layer are 40, and the neuron number of output layer is 1.Choosing
With purline as output function, the sentence of setting up of network is:
Bpnet=newff (minmax (P), [41], ' tansig', ' purelin'}, ' traingdx', '
learngdm') (29)
It is the discrete signal after NN filtering to export content, and it is anti-that algorithm carries out error using steepest descent method
To propagation, the successive ignition by neutral net can obtain the BP neural network with filter function for mixed noise.
Next generalized correlation method is used, suitable weight function is chosen and carries out peakvalue's checking, the time delay estimation so obtained is more smart
It is accurate.Finally by formula 3, you can determine the position of leakage point, Fig. 3 show the structural model of neutral net positioning mode.Right
The 500HZ that 5000HZ is brought into by the generation high frequency out-of-band noise such as bubble explosion, water impact medium and sensor electric signal is low
After out-of-band noise processing, in addition it is also necessary to prevent the noise jamming with frequency range with water leakage.Now just filtered using through band logical
Data after ripple processing as neutral net input and set up BP neural network to carry out leak source positioning.The system uses three layers
The neuron node number of the BP neural network of structure, wherein input layer is 1, and object vector is to believe after earlier data is handled
Number sample sequence, be to obtain under experimental conditions;Its input vector be actual conditions in contain various unknown noises in the case of
Signal sample sequence.The conventional excitation function of the hidden layer of network includes tansig and logisg, by network training, for
Same anticipation error is reached, tansig training pace is less than logsig, therefore here from tansig functions as sharp
Function is encouraged, it is 40 that the multiple training test by neutral net, which obtains node in hidden layer, the neuron number of output layer is 1,
Error back propagation is carried out using steepest descent method, the neutral net trained by successive ignition passes through actual conditions
Carry out the feasibility of verification method.
During time delay estimation is carried out using neutral net, its process is divided into two stages:First stage is to learn
Habit process, second stage is the course of work, is comprised the following steps that:
1) sample data construction training sample is chosen.Because water supply line is buried in underground, its local environment is complicated and changeable,
It is a unstable nonlinear system, it is therefore necessary to select the sample data under home, the unusual sample being otherwise drawn into
This can reduce the predictive ability of network.Sample data is pre-processed., be to original number before neural network prediction is carried out
According to being normalized, make its data standard between [- 1,1].
2) training sample is constructed.The signal sample sequence after noise suppressed is as object vector under experimental conditions, so
The water leakage containing different noises is measured in practical situations both afterwards, obtained sample sequence is as input signal.
3) BP neural network for choosing three-decker sets up forecast model.The corresponding node of input layer, hidden layer, output layer
Number is respectively 1,40 and 1, and hidden layer output layer functions are respectively tansig and purelin functions.
4) training network.Before training network, in addition it is also necessary to set training parameter, wherein maximum training pace
Net.trainParam.epochs is 10000, and least mean-square error net.trainParam.goal is 0.05, and study speed
Rate is set to 0.01 etc..
After neural network completion, the water leakage measured is carried out in band, out-of-band noise suppresses.The present invention is carried
The water supply line leak detection location model gone out is totally divided into two parts:Leak detection is carried out first, is sent out when detecting leakage
After life, followed by the positioning of leakage point.Its overall framework as shown in figure 4, test result indicates that:No matter which kind of noise is used
Estimation mode, spectrum-subtraction can lift the signal to noise ratio of water leakage;And use " the No leakage estimation technique " to estimate noise spectrum to noise
The lifting of ratio is the most obvious.Next when being verified to Spectral variance algorithm simulating, the Spectral variance of noise model is calculated first,
Double threshold threshold value T1 and T2 are determined, then calculates the Spectral variance that spectrum subtracts each signal frame after enhancing.As shown in figure 5,0.3 second it
It is preceding and 3.6 seconds after be the quiet section for not leaking generation.Three width image as shown in Figure 5 is non-superimposed noise, signal to noise ratio respectively
It is to calculate obtained short-term spectrum variance image under -5dB, three kinds of background noise conditions for 5dB and signal to noise ratio.Parallel to horizontal stroke in figure
The dotted line of axle is threshold value T1 image, and solid line is threshold value T2 image.Parallel to solid marks the opening for leakage signal of the longitudinal axis
Top, dashed lines labeled is that leakage signal terminates end.From image, the signal detecting method based on spectrum-subtraction and Spectral variance is not
Only there is good separating capacity in the case of high s/n ratio, still there is powerful performance in the case of low signal-to-noise ratio.Even in letter
When making an uproar than for -5dB, occur once judging by accident only at 2.7 seconds, but do not influence the validity of overall algorithm, leakage signal section and the back of the body
Still difference is obvious for the waveform of scape noise segment.
Fig. 6 show the signal amplitude after normalization.The Spectral variance of signal shown in Fig. 6 is calculated, is as a result shown, in input
10 gradients that Signal-to-Noise changes from -5dB to 5dB, the Spectral variance value maximum of noise model is no more than 30, and strengthens
The Spectral variance value of water leakage is more than 680 afterwards.The two difference is obvious, by setting suitable threshold value can be with effective detection
Go out leakage signal.
In summary, data are either still calculated from emulating image and reflect letting out based on spectrum-subtraction and Spectral variance
Substantially, validity is strong for leakage signal detection method Detection results., it is necessary to carry out accurately determining to leak after leak detection is completed
Position.It is respectively the waveform and frequency spectrum and the feelings after bandpass filtering of the water leakage measured rush-hour as shown in Figure 7
Condition.The water leakage of rush-hour is being obtained, out-of-band noise can be reduced by bandpass filtering, it is right after so processing
Resulting signal is sampled, and the matrix that sampled point is constituted is as input vector, the letter when night noise is smaller to measuring
Number sampled, the matrix of composition is trained, training is as shown in Figure 8 as object vector to neutral net.Obviously,
0.64% error precision has been reached by 9991 training, next another section of water leakage of rush-hour has been regard as survey
Examination input, output obtains the signal filtered by BP neural network.By identical step process sensor 1,2 in rush-hour
Water leakage, time delay is next tried to achieve using generalized correlation method.
First have to select weight function during pre-filtering is carried out, common weight function includes:Substantially it is related, at Roth
Reason, the smooth coherence transfers of SCOT, PHAT phse conversions, Eckart, this 6 kinds of ML maximum likelihoods weight function.When selecting weight function,
It is ensured that have a spike rather than a broad peak in cross-correlation function, and have between high-resolution and stability one it is simultaneous
Turn round and look at.It is next just different respectively after having carried out feasibility analysis to BP neural network and generalized correlation in terms of leak positioning
Noise type, signal to noise ratio, the weighting function of generalized correlation method and neural network parameter these four variables to arithmetic accuracy
Influence is analyzed, and then draws full experiment conclusion.Influence of the different noise types to algorithm.Ensureing that its dependent variable is identical
In the case of, the signal noise being mixed into successively be white Gaussian noise, coloured noise both carry out arithmetic results comparison.First
White noise is mixed into, its training error such as Fig. 9 (a) is shown, then the signal of the sensor 1 and 2 after NN filtering is carried out
Generalized correlation, shown in obtained cross-correlation function image such as Fig. 9 (b).Next frequency is mixed into make an uproar in the coloured of 500-2000HZ
Equivalent to increase in-band noise after sound, the broad sense cross-correlation function of its neural metwork training result and two sensorses is respectively such as Figure 10
In shown in (a) and 10 (b).
In the case where being mixed into white noise, neural metwork training precision reaches anticipation error 0.001 through 1599 steps, and is mixed into
The training precision of coloured noise only reaches 0.0024 through 10000 steps, illustrates that the network training precision containing white noise is better than containing coloured
The situation of noise.
Influence of the different signal to noise ratio to arithmetic accuracy, exemplified by being mixed into white Gaussian noise, in the range of 10dB to -12dB
Change signal to noise ratio, let the signal go through the error curve obtained under BP neural network, wherein 10dB and -12dB as shown in figure 11, will
Error result under multiple signal to noise ratio is arranged into following table, it can be found that signal to noise ratio is higher, the training result of neutral net is unreasonable
Think, it is more accurate using the positioning of generalized correlation.
Following table is the network training result under different signal to noise ratio
Algorithm parameter setting itself can also be impacted to positioning precision.The setting of BP algorithm parameter mainly includes hidden layer
These aspects of nodes, transfer function, training method.Research object is set to the leak that signal to noise ratio containing white Gaussian noise is 10dB and believed
Number, it is trained first in the network containing different node in hidden layer, if maximum step-length is 10000 steps, changes hidden layer
Nodes and carry out repeatedly training and obtain the average of training precision, when training sample is 5000, training result is as shown in the table,
The optimal nodes that hidden layer can be obtained are 17, and now training error precision is minimum and is 0.00234, as a result shows hidden layer
Nodes have an impact on error precision but influence smaller, and the increase of nodes is bigger for operation time and operand influence.
Following table is hidden layer node and the training precision table of comparisons of neutral net
And for transfer function, the conventional transfer function of hidden layer is logsig and tansig, for same letter
Number, in identical maximum step-length, tansig training precision is higher than logsig, so selection tansig is used as hidden layer
Transfer function.The weight function of generalized correlation method also has an impact to arithmetic accuracy, and common weight function has 5 kinds, along with traditional phase
Pass method (CC), altogether 6 kinds of processing methods.Different methods are respectively adopted generalized correlation processing is carried out to same water leakage, when
The signal to noise ratio snr 1 of sensor 1 is 10dB, when the signal to noise ratio snr 2 of sensor 2 is 5dB, its different generalized related function image
As shown in figure 12.
When the signal to noise ratio snr 1 of sensor 1 is 10dB, when the signal to noise ratio snr 2 of sensor 2 is -5dB, its different broad sense
Correlation function image is as shown in figure 13, it can be seen that as SNR1=10dB and SNR2=-5dB, with the decline of signal to noise ratio, before
Correlation function peak value more sharp SCOT and ROTH has been submerged in noise, and CC, PHAT peak value also become gentle therewith
And it is vulnerable to the influence of secondary peak.Reviewing ML and Eckart peak value sharpness does not have notable change with the decline of signal to noise ratio
Change.In summary, when selecting generalized related function to carry out leak positioning, when the signal to noise ratio that sensor receives signal is higher
When, select the locating effect of PHAT and SCOT weight functions preferable;When its signal to noise ratio is relatively low, selection ML and Eckart weight functions
Locating effect is preferable.
One kind of embodiment described above, simply more preferably embodiment of the invention, those skilled in the art
The usual variations and alternatives that member is carried out in the range of technical solution of the present invention all should be comprising within the scope of the present invention.
Claims (11)
1. a kind of water supply line leakage detecting and locating method, it is characterised in that the signal that methods described is gathered to sensor is carried out
Enhancing, is then calculated the Spectral variance of signal after enhancing, is judged using double threshold method, if Spectral variance is in threshold range
Inside then there is leakage in explanation, then carry out leak point location, wave filter be constituted using BP neural network, by water leakage from noise
Separation, generalized correlation is carried out to water leakage, is selected the good weight function of performance according to signal to noise ratio, is carried out time delay estimation, obtain three
Individual sensor when delay, calculated using leak location model, obtain leak dot position information.
2. according to the method described in claim 1, it is characterised in that methods described includes:
S1:Water leakage progress spectrum is subtracted into enhancing;
S2:Calculate the Spectral variance of signal after strengthening in S1;
S3:Leakage signal is recognized using one-parameter double-threshold comparison, and determines threshold value;
S4:Leak detection is carried out, and exports testing result;
S5:To the testing result exported in S4, out-of-band noise is removed by bandpass filtering;
S6:NN filtering is carried out, in-band noise is removed;
S7:Carry out generalized correlation for time delay estimation;
S8:Using the time delay estimated result in S7, by leak location model, leak dot position information is obtained, leak is completed and determines
Position.
3. method according to claim 2, it is characterised in that the S7 is specially:During selection of weighting function, need
To select the good weight function of performance to carry out time delay estimation according to signal to noise ratio, obtain three sensors when delay, utilize the leakage
Water location model is calculated, and obtains leak dot position information.
4. method according to claim 3, it is characterised in that the leak location model is three sensor location models,
The model equidistantly distributes sensor around water supply network, wherein the three sensor nodes numbering nearest apart from leak source
Respectively 1,2,3, the distance between they are L, and sensor 1,2 is set to d1 and d2 apart from the position of leak source, passes through correlation
The time delay that method is measured between sensor 1 and 2,2 and 3 is respectively D12And D23, the spread speed ν where water leakage on pipeline:ν
=L/D23, distance of the leak source away from sensor 1 and 2 and it is respectivelyWith
5. method according to claim 2, it is characterised in that the S1 is specially:
S11:Input water leakage s (n) is pre-processed, the pretreatment includes preemphasis and adding window framing;
S12:The leakage signal made an uproar to band carries out Short Time Fourier Analysis, calculates the short-time energy spectrum for obtaining each frame signal | Ym
(ω)|2;
S13:Before strengthening using spectrum-subtraction water leakage, noise spectrum estimation is carried out, noise spectrum estimation value is obtainedWith | Ym(ω)|2SubtractSo as to obtain removing the leakage signal power spectrum after Noise enhancement
S14:Spectrum amplitude value to obtaining leakage signal after result sqrt in S13With reference to the phase of former water leakage
Position information obtains the spectrum estimation of each frame leakage signalCarry out inverse Fourier transform again, to leakage signal carry out recover and
Reconstruct, obtains spectrum and subtracts enhanced leakage signal.
6. method according to claim 2, it is characterised in that the calculating of signal spectrum variance is specific as follows in the S2:
S21:Assuming that input signal is s (n), the length per frame is N, and signal is transformed from the time domain into frequency-domain calculations frequency by DFT
Spectrum:
Record each frequency component with a matrix | S (ω) | value;
S22:Calculate the average of each component:
S23:Calculate the Spectral variance value of enhanced leakage signal in previous step:And obtain
The average value of noise model Spectral variance
7. method according to claim 2, it is characterised in that the S3 is specially:
S31:Two threshold values T1 and T2 are set, the average value of noise model Spectral variance is obtained
S32:Threshold value T1 is set toThreshold value T2 is set to。
8. method according to claim 2, it is characterised in that the S4 is specially:
S41:It is leakage signal when parameter D is higher than threshold value T2;
S42:T1 positions are higher or lower than by D, to judge the terminal of leakage signal;
S43:Statistics, if frame number is more than a quarter of input signal totalframes, is determined as higher than the signal frame number of threshold value
Leakage, and output signal, are otherwise No leakage.
9. method according to claim 2, it is characterised in that the process being filtered in the S6 using neutral net
In, its process is divided into two stages:First stage is learning process, and second stage is the course of work.
10. method according to claim 9, it is characterised in that the learning process is:
1) sample data under selection leakage environment is used as training sample;
2) sample data is pre-processed, before neural network prediction is carried out, initial data is normalized, made
Its data standard is between [- 1,1];
3) training sample is constructed, the signal sample sequence after noise suppressed is as object vector under experimental conditions, in reality
In the case of measure the water leakage containing different noises, obtained sample sequence is as input signal.
11. method according to claim 9, it is characterised in that the course of work is:
A) BP neural network for choosing three-decker sets up forecast model;The corresponding interstitial content of input layer, hidden layer, output layer
Respectively 1,40 and 1, hidden layer, output layer functions are respectively tansig and purelin functions;
B) training network;Before training network, in addition it is also necessary to set training parameter, wherein maximum training pace
Net.trainParam.epochs is 10000, and least mean-square error net.trainParam.goal is 0.05, and study speed
Rate is set to 0.01 etc.;
C) after neural network completion, the water leakage collected is carried out in band, out-of-band noise suppresses.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710139772.7A CN107061996B (en) | 2017-03-10 | 2017-03-10 | A kind of water supply line leakage detecting and locating method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710139772.7A CN107061996B (en) | 2017-03-10 | 2017-03-10 | A kind of water supply line leakage detecting and locating method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107061996A true CN107061996A (en) | 2017-08-18 |
CN107061996B CN107061996B (en) | 2019-04-16 |
Family
ID=59621679
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710139772.7A Active CN107061996B (en) | 2017-03-10 | 2017-03-10 | A kind of water supply line leakage detecting and locating method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107061996B (en) |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107869653A (en) * | 2017-12-14 | 2018-04-03 | 北京化工大学 | A kind of pipeline flow sensitivity matrix soil's rigidity method |
CN108286655A (en) * | 2017-12-28 | 2018-07-17 | 天津工业大学 | A kind of water flow Photoelectric Detection and intelligent water leakage method of discrimination and equipment therefor |
CN108443714A (en) * | 2018-04-09 | 2018-08-24 | 山东省科学院激光研究所 | Determine the method and device of pipe leakage |
CN108758356A (en) * | 2018-05-20 | 2018-11-06 | 东北石油大学 | A kind of information processing method and system |
CN108758354A (en) * | 2018-05-03 | 2018-11-06 | 太原理工大学 | Heat supply pipeline leak detection system and method based on infrasound and reference point |
CN109451529A (en) * | 2019-01-02 | 2019-03-08 | 成都华日通讯技术有限公司 | A kind of miniature spectral sensor net system and its synergistic data processing method |
CN109443669A (en) * | 2018-09-30 | 2019-03-08 | 深圳市英威腾电源有限公司 | A kind of leakage detection method and device |
CN109555977A (en) * | 2018-11-23 | 2019-04-02 | 水联网技术服务中心(北京)有限公司 | The equipment and recognition methods of leak noise measuring |
CN109555979A (en) * | 2018-12-10 | 2019-04-02 | 清华大学 | A kind of water supply network leakage monitoring method |
CN109681789A (en) * | 2019-03-01 | 2019-04-26 | 重庆邮电大学 | The relevant detection localization method of fluid line leak acoustic signal based on variation mode decomposition |
CN110136210A (en) * | 2018-02-07 | 2019-08-16 | 中国石油化工股份有限公司 | A kind of intelligent early-warning system of monitoring device leakage |
CN110195823A (en) * | 2019-07-16 | 2019-09-03 | 重庆科技学院 | A kind of micro- leakage detection method of pipeline based on deep neural network |
CN110261821A (en) * | 2019-07-18 | 2019-09-20 | 中电科仪器仪表有限公司 | One kind demodulating relevant time difference estimation method and device based on information |
CN110529745A (en) * | 2019-08-05 | 2019-12-03 | 上海第二工业大学 | The algorithm of fiber laser arrays pipe leakage and position under a kind of frequency domain |
CN110953485A (en) * | 2019-12-06 | 2020-04-03 | 北京无线电计量测试研究所 | Gas pipeline leakage point positioning method and system |
CN111365625A (en) * | 2020-04-13 | 2020-07-03 | 安徽理工大学 | Pipeline leakage sound wave positioning method based on delay summation |
CN111425768A (en) * | 2020-03-31 | 2020-07-17 | 长安大学 | Oil pipeline oil leakage point and oil leakage rate detection method based on underwater sensor network |
CN111895278A (en) * | 2020-09-03 | 2020-11-06 | 武汉轻工大学 | Urban water supply pipe network leakage point positioning method, device, equipment and storage medium |
CN112013285A (en) * | 2020-08-26 | 2020-12-01 | 辽宁石油化工大学 | Method and device for detecting pipeline leakage point, storage medium and terminal |
CN112992178A (en) * | 2021-02-05 | 2021-06-18 | 安徽绿舟科技有限公司 | Method for analyzing pipeline leakage through chaotic voiceprint |
WO2021208521A1 (en) * | 2020-04-13 | 2021-10-21 | 安徽理工大学 | Pipeline leakage position calculation method based on beam forming |
CN113780381A (en) * | 2021-08-28 | 2021-12-10 | 特斯联科技集团有限公司 | Artificial intelligence water leakage detection method and device |
CN114234061A (en) * | 2021-12-20 | 2022-03-25 | 北京工业大学 | Neural network-based intelligent judgment method for water leakage sound of pressurized operation water supply pipeline |
CN114738679A (en) * | 2022-04-11 | 2022-07-12 | 长江大学 | Pipeline missing detection device and pipeline missing detection method |
CN116293486A (en) * | 2023-03-15 | 2023-06-23 | 东莞先知大数据有限公司 | Water pipe leakage detection method and device, electronic equipment and storage medium |
CN116464918A (en) * | 2023-05-06 | 2023-07-21 | 江苏省特种设备安全监督检验研究院 | Pipeline leakage detection method, system and storage medium |
CN116577037A (en) * | 2023-07-12 | 2023-08-11 | 上海电机学院 | Air duct leakage signal detection method based on non-uniform frequency spectrogram |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110159868B (en) * | 2019-05-27 | 2021-07-30 | 北京奥蓝仕技术有限公司 | Data processing method and device for intelligent sleeve and intelligent sleeve |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0714019B1 (en) * | 1994-11-22 | 1999-09-22 | Gaz De France (Service National) | Procedure and device for the detection and localisation of leaks of fluid flowing in a pipe |
DE202008017243U1 (en) * | 2008-05-02 | 2009-09-17 | Rbs Wave Gmbh | System for monitoring and / or locating water losses in a water pipeline network |
CN101551064A (en) * | 2009-05-22 | 2009-10-07 | 重庆大学 | Water supply pipe leakage detection locating signal processing method |
CN102193082A (en) * | 2011-03-18 | 2011-09-21 | 哈尔滨工业大学 | Device for positioning leak source of three-sensor multi-scale constrained pipe network |
CN104061445A (en) * | 2014-07-09 | 2014-09-24 | 中国石油大学(华东) | Pipeline leakage detection method based on neural network |
-
2017
- 2017-03-10 CN CN201710139772.7A patent/CN107061996B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0714019B1 (en) * | 1994-11-22 | 1999-09-22 | Gaz De France (Service National) | Procedure and device for the detection and localisation of leaks of fluid flowing in a pipe |
DE202008017243U1 (en) * | 2008-05-02 | 2009-09-17 | Rbs Wave Gmbh | System for monitoring and / or locating water losses in a water pipeline network |
CN101551064A (en) * | 2009-05-22 | 2009-10-07 | 重庆大学 | Water supply pipe leakage detection locating signal processing method |
CN102193082A (en) * | 2011-03-18 | 2011-09-21 | 哈尔滨工业大学 | Device for positioning leak source of three-sensor multi-scale constrained pipe network |
CN104061445A (en) * | 2014-07-09 | 2014-09-24 | 中国石油大学(华东) | Pipeline leakage detection method based on neural network |
Non-Patent Citations (2)
Title |
---|
张建利等: "三传感器约束多尺度相关管道漏点定位方法", 《哈尔滨工业大学学报》 * |
王路露等: "基于频谱方差和谱减法的语音端点检测新算法", 《计算机工程与应用》 * |
Cited By (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107869653B (en) * | 2017-12-14 | 2019-12-06 | 北京化工大学 | Pipeline flow sensitivity matrix leakage detection method |
CN107869653A (en) * | 2017-12-14 | 2018-04-03 | 北京化工大学 | A kind of pipeline flow sensitivity matrix soil's rigidity method |
CN108286655A (en) * | 2017-12-28 | 2018-07-17 | 天津工业大学 | A kind of water flow Photoelectric Detection and intelligent water leakage method of discrimination and equipment therefor |
CN108286655B (en) * | 2017-12-28 | 2024-03-15 | 天津工业大学 | Intelligent water flow photoelectric detection and water leakage judging method and device |
CN110136210B (en) * | 2018-02-07 | 2023-06-09 | 中国石油化工股份有限公司 | Intelligent early warning system for leakage of monitoring device |
CN110136210A (en) * | 2018-02-07 | 2019-08-16 | 中国石油化工股份有限公司 | A kind of intelligent early-warning system of monitoring device leakage |
CN108443714A (en) * | 2018-04-09 | 2018-08-24 | 山东省科学院激光研究所 | Determine the method and device of pipe leakage |
CN108758354B (en) * | 2018-05-03 | 2023-09-12 | 太原理工大学 | Heat supply pipeline leakage detection system and method based on infrasonic wave and reference point |
CN108758354A (en) * | 2018-05-03 | 2018-11-06 | 太原理工大学 | Heat supply pipeline leak detection system and method based on infrasound and reference point |
CN108758356A (en) * | 2018-05-20 | 2018-11-06 | 东北石油大学 | A kind of information processing method and system |
CN109443669A (en) * | 2018-09-30 | 2019-03-08 | 深圳市英威腾电源有限公司 | A kind of leakage detection method and device |
CN109555977A (en) * | 2018-11-23 | 2019-04-02 | 水联网技术服务中心(北京)有限公司 | The equipment and recognition methods of leak noise measuring |
CN109555979A (en) * | 2018-12-10 | 2019-04-02 | 清华大学 | A kind of water supply network leakage monitoring method |
CN109451529A (en) * | 2019-01-02 | 2019-03-08 | 成都华日通讯技术有限公司 | A kind of miniature spectral sensor net system and its synergistic data processing method |
CN109681789B (en) * | 2019-03-01 | 2020-07-03 | 重庆邮电大学 | Fluid pipeline leakage acoustic vibration coherent detection positioning method based on variational modal decomposition |
CN109681789A (en) * | 2019-03-01 | 2019-04-26 | 重庆邮电大学 | The relevant detection localization method of fluid line leak acoustic signal based on variation mode decomposition |
CN110195823A (en) * | 2019-07-16 | 2019-09-03 | 重庆科技学院 | A kind of micro- leakage detection method of pipeline based on deep neural network |
CN110261821A (en) * | 2019-07-18 | 2019-09-20 | 中电科仪器仪表有限公司 | One kind demodulating relevant time difference estimation method and device based on information |
CN110529745A (en) * | 2019-08-05 | 2019-12-03 | 上海第二工业大学 | The algorithm of fiber laser arrays pipe leakage and position under a kind of frequency domain |
CN110953485A (en) * | 2019-12-06 | 2020-04-03 | 北京无线电计量测试研究所 | Gas pipeline leakage point positioning method and system |
CN111425768A (en) * | 2020-03-31 | 2020-07-17 | 长安大学 | Oil pipeline oil leakage point and oil leakage rate detection method based on underwater sensor network |
WO2021208521A1 (en) * | 2020-04-13 | 2021-10-21 | 安徽理工大学 | Pipeline leakage position calculation method based on beam forming |
CN111365625A (en) * | 2020-04-13 | 2020-07-03 | 安徽理工大学 | Pipeline leakage sound wave positioning method based on delay summation |
CN112013285B (en) * | 2020-08-26 | 2022-04-19 | 辽宁石油化工大学 | Method and device for detecting pipeline leakage point, storage medium and terminal |
CN112013285A (en) * | 2020-08-26 | 2020-12-01 | 辽宁石油化工大学 | Method and device for detecting pipeline leakage point, storage medium and terminal |
CN111895278A (en) * | 2020-09-03 | 2020-11-06 | 武汉轻工大学 | Urban water supply pipe network leakage point positioning method, device, equipment and storage medium |
CN112992178A (en) * | 2021-02-05 | 2021-06-18 | 安徽绿舟科技有限公司 | Method for analyzing pipeline leakage through chaotic voiceprint |
CN113780381A (en) * | 2021-08-28 | 2021-12-10 | 特斯联科技集团有限公司 | Artificial intelligence water leakage detection method and device |
CN113780381B (en) * | 2021-08-28 | 2022-07-01 | 特斯联科技集团有限公司 | Artificial intelligence water leakage detection method and device |
CN114234061A (en) * | 2021-12-20 | 2022-03-25 | 北京工业大学 | Neural network-based intelligent judgment method for water leakage sound of pressurized operation water supply pipeline |
CN114738679A (en) * | 2022-04-11 | 2022-07-12 | 长江大学 | Pipeline missing detection device and pipeline missing detection method |
CN114738679B (en) * | 2022-04-11 | 2023-09-22 | 长江大学 | Pipeline missing detection device and pipeline missing detection method |
CN116293486A (en) * | 2023-03-15 | 2023-06-23 | 东莞先知大数据有限公司 | Water pipe leakage detection method and device, electronic equipment and storage medium |
CN116464918B (en) * | 2023-05-06 | 2023-10-10 | 江苏省特种设备安全监督检验研究院 | Pipeline leakage detection method, system and storage medium |
CN116464918A (en) * | 2023-05-06 | 2023-07-21 | 江苏省特种设备安全监督检验研究院 | Pipeline leakage detection method, system and storage medium |
CN116577037B (en) * | 2023-07-12 | 2023-09-12 | 上海电机学院 | Air duct leakage signal detection method based on non-uniform frequency spectrogram |
CN116577037A (en) * | 2023-07-12 | 2023-08-11 | 上海电机学院 | Air duct leakage signal detection method based on non-uniform frequency spectrogram |
Also Published As
Publication number | Publication date |
---|---|
CN107061996B (en) | 2019-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107061996B (en) | A kind of water supply line leakage detecting and locating method | |
CN103854662B (en) | Adaptive voice detection method based on multiple domain Combined estimator | |
CN107991706B (en) | Coal seam hydraulic fracturing microseismic signal combined noise reduction method based on wavelet packet multiple threshold and improved empirical mode decomposition | |
CN103308804B (en) | Based on quick K-S converting electric power quality disturbance signal time and frequency parameter extracting method | |
CN110456332B (en) | Underwater acoustic signal enhancement method based on automatic encoder | |
CN112802484B (en) | Panda sound event detection method and system under mixed audio frequency | |
CN103632676B (en) | A kind of low signal-to-noise ratio voice de-noising method | |
CN103675617A (en) | Anti-interference method for high-frequency partial discharge signal detection | |
CN113094993B (en) | Modulation signal denoising method based on self-coding neural network | |
CN105424366A (en) | Bearing fault diagnosis method based on EEMD adaptive denoising | |
CN113378661A (en) | Direct current electric energy signal denoising method based on improved wavelet threshold and related detection | |
CN102419972B (en) | Method of detecting and identifying sound signals | |
CN110221256A (en) | SAR disturbance restraining method based on depth residual error network | |
CN113642484B (en) | Magnetotelluric signal noise suppression method and system based on BP neural network | |
CN112183225A (en) | Underwater target signal feature extraction method based on probability latent semantic analysis | |
Yao et al. | An adaptive seismic signal denoising method based on variational mode decomposition | |
CN107368799A (en) | Leakage detection and localization method based on multiple features and adaptive time delay stimation | |
CN114242099A (en) | Speech enhancement algorithm based on improved phase spectrum compensation and full convolution neural network | |
CN111580151A (en) | SSNet model-based earthquake event time-of-arrival identification method | |
Kimiaefar et al. | Random noise attenuation by Wiener-ANFIS filtering | |
CN113472390A (en) | Frequency hopping signal parameter estimation method based on deep learning | |
Guo et al. | LFM signal optimization time-fractional-frequency analysis: Principles, method and application | |
CN111650655B (en) | Non-negative matrix factorization supervised transient electromagnetic signal noise reduction method | |
CN113822363B (en) | Convolutional neural network time delay estimation method based on quadratic rational kernel function | |
CN107635181B (en) | Multi-address sensing source feedback optimization method based on channel learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |